Award
NIH Reporter #5R01EY036092-02
The Development of Clinically-Motivated and Explainable Machine-Learning-Based Image Classifiers to Detect Trachoma
Recipient
University of Vermont & St. College, Burlington, VT, United States
Award Amount
$461,250.00
Ceiling
$461,250.00
Awarded
January 30, 2026
Identifier
5R01EY036092-02
Development of AI-based image classifiers on smartphones to detect trachoma, supporting global elimination efforts and enabling remote diagnosis without cloud dependency.
Description
This project proposes to develop a smartphone-based application to standardize image acquisition, identify ungradable images, and improve the performance of machine learning models to detect trachoma features such as follicles, inflammation, and scarring. The models will be deployed offline on Android devices to facilitate remote diagnosis without cloud dependency. The research aims to improve diagnostic accuracy, support global elimination efforts, and generate new knowledge about the active trachoma phenotype.